The Ladder of Evidence: Get More Value From Your Customer Interviews and Product Experiments

Teresa Torres introduces the Ladder of Evidence, a framework for judging how reliable customer feedback is. Asking people what they would do in the future, as focus groups and surveys often do, yields weak signal because speculation about future behavior is unreliable. Climbing the ladder means moving from hypothetical questions to past behavior, then to specific stories about past behavior, which carry rich context and emotion. For testing solutions, the top rungs favor observing action over opinion: simulating the real experience, or watching people use a competitor's product. The argument matters because it helps teams gather trustworthy evidence quickly without either over-researching or building the wrong thing.

01Key takeaways

• Treat hypothetical 'would you use this?' answers as unreliable and avoid relying on them.
• Ask customers about specific past behavior rather than what they generally do or might do.
• Collect detailed stories to uncover context, emotion, and the multiple motivations behind behavior.
• Test solutions by simulating the real experience and measuring actual action.
• Start with the cheapest simulation that still reveals behavior, balancing learning speed against build cost.
• When a competitor exists, observe how people use their product to validate your idea sooner.

02Key sections

Why speculative questions fail

Asking people whether they would use or pay for something invites easy but unreliable answers. Future intentions often diverge from actual behavior.

Climbing from facts to stories

Past behavior predicts future behavior better than speculation, and specific stories reveal context, emotion, and motivations that bare facts miss.

Testing by simulating experience

Instead of asking whether people like a concept, create a realistic experience and measure whether they take action, such as attending a pilot session.

Using competitor products

When building something similar to an existing offering, observe users interacting with the competitor's product, either as a task demo or in real-life use.

03From the post

“One of the best signs of an effective product team is a regular cadence of customer interviews and product experiments. But it’s not enough to simply check the box on these activities; we need to make sure that we do them effectively. Without a strong background in research methods,”

04Frameworks mentioned

Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.